You need to read both BUG-REPORTING and MAINTAINERS :)
For bug reports, e-mail the kernel eepro100 maintainer too:
EEPRO100 NETWORK DRIVER
P: Andrey V. Savochkin
M: [EMAIL PROTECTED]
S: Maintained
When reporting bugs, you absolutely must report the kernel version and
hardware chip version which you are using. Additionally, since the
kernel drivers are often different from Donald Becker's web site
drivers, you should report whether you are using a Becker driver or a
derivative.
Jeff
Dennis wrote:
>
> I emailed this 3 weeks ago without response. I think it is rather important:
>
> >I have identified a few existing problems with the eepro100 v.1.09t driver:
> >
> >1) speedo_show_stats will crash if a ringp[i] receive ring entry is null
> >2) the driver does not recover from a situation where the receiver cannot
> replace multiple a receive sk_buff (related to above).
> >
> >There doesnt seem to be a working cleanup procedure to replenish failed
> buffer allocation ring entries under extreme conditions
> >
> >After fixing 1), after a transmit timeout all receive on the interface
> ceases (although network buffers in use is down to 38 or so, indicating
> that buffers are available).
> >
> >ifconfig eth1 down
> >ifconfig eth1 up
> >
> >restores operation.
> >
> >This can be created by having a system with a 100Mb/s interface and a
> 10Mb/s and flooding the 100Mb/s interface with traffic...I think that the
> receive queue grows until memory is exhausted (another separate issue
> perhaps)....
> >
> >Let me know if you have any simple fixes for it please.
> >
> >Dennis Baasch
> >
--
Jeff Garzik | Tact is the ability to tell a man
Building 1024 | he has an open mind when he has a
MandrakeSoft, Inc. | hole in his head. (-random fortune)
-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to [EMAIL PROTECTED]
